OT: A "decomposed" business structure

That can be solved with progress reports and milestones. Maybe better, because everyone involved can share the results, and they can be archived.

Leadership. It's easier to get people fired up about something when you can get them together in front of you. Hard, maybe impossible, to do remotely. Without excellent leadership, things are going to go to pot eventually. If you select for just those folks who can self-lead you'll probably filter out a lot of really good people (and perhaps end up with more obstreperous curmudgeons that prefer not to work with other prople directly). How do you have a heated argument over e-mail and come to some kind of resolution that works? (The first part is easy, the second, I think less easy).

There are a few places where definite physical location is required- the gov't will want to know where the books and records for the corp are located, as well as the exact location of any 'special' goods they deem worth tracking. I can't imagine getting a bank account without an address. Those could be a contractor or whatever, I suppose. It would complicate payroll functions, for example, if people are in different jurisdictions. If different countries are involved, there will be duplication involved unless everyone is an independent contractor.

Pieces of it. My overall impression is that some people seem to be happier, but it's not as effective as offices. I think it can work for a time, in some situations. I did an entire mid-sized project with a fully distributed company, but we distributed folks all knew and trusted each other, and we were all very competent in our (well-defined) fields. I don't think we were ever all in the same room, and the first time I met a few of the group was on the aircraft to meet with the client.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Remote oursourcing of monolithic functions isn't unusual in large organisations. I've worked with outfits where the entire accounting/finance division is in India or China, that's actually not uncommon. Setting it up successfully demands good formal internal procedures and standards before starting out, smaller organisations may be less likely to have them in place.

I regularly arrange chunks of engineering to be done by divisions overseas that I've never met. It's surprising how quickly a strong, warm relationship can develop, even when the only contact is via email and phone. Along with the good relationship tends to come a quality outcome. However, if things go off the rails for any reason, it's a lot harder to sort things out without face to face contact. I prefer to only do remote stuff if I know I can cope with a total bombout.

It's amazing how complex, misunderstood, festering technical problems, things that have evoked scores of emails and PowerPoints, can get resolved by getting people physically together in front of a whiteboard. For something complex like electronic design, the distributed model sounds dangerous to me. It's rare to find people who work well this way, and tekkies are often not good writers, in that they don't always express themselves clearly in emails or even on the telephone. We see this all the time.

Outsourcing PCB layout is just one example of how distributed engineering can be difficult. Ditto embedded code.

Brainstorming is fundamental to the early stages of product development. I think that requires physical presence.

There have been successful companies that had scores of small divisions. EG&G, Perkin-Elmer, Vishay, Bruker for example.

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

Or in front of a bench and an oscilloscope or logic analyzer.

I don't know how many times I've seen a technical problem solved by through the simple act of attempting to demonstrate and explain the problem to somebody else. You just can't do that via email or even on the phone.

--
Grant

[snip]

Well, without tipping my hand too much, I do. And one of the major issues I've encountered is that the various regulatory and taxing authorities are going to shit themselves if there isn't an actual physical site they can kick the doors in on.

In Washington State, for example, you cannot have a corporation, partnership, or whatever without a physical address. So I work as an "employee" of a foreign firm. They haven't gotten around to as asking for that address. Yet.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
I have a very firm grasp on reality. I can reach out and strangle it any
time!

Yes. Though some activities don't really lend themselves easily to these sorts of mechanisms (imagine the accountant's role -- aside from things like: have the payroll ready for deposit to accounts by 11PM thursday).

One thing I thought might help is a whiteboard sort of approach where everyone's activities are visible to everyone else. This has some distinct advantages -- if you're not doing your job, folks will tend to see it sooner; if you take ill, there is a path that another can try to pick up (hmmm... looks like he was getting ready to file these license forms...); it let's folks see how you (mis)*interpreted* past decisions; etc.

Video conferencing. Don't restrict yourself to 1960's communication technology...

I don't see how physical presence makes a difference given what is now readily affordable in that regard. One "must have" (IMO) is a shared whiteboard application where each party can have a tablet and share ink+boardspace. I.e., I can scribble on *my* "whiteboard" and *you* can augment/modify it concurrently. Prepare new drawings off-line (before/during/after such a conference) and "reveal" them as necessary during your discussion.

This gives you the shared workspace that is available in a "conference room", allows "presentations" to be seemlessly and interactively shared and documents the entire process (for later playback/review).

I.e., if you come up with an idea *later*, you can amend a "discussion/presentation" to illustrate how your idea dovetails with the original concept -- or, takes it in a different direction.

These are "easy" things to address. And, depend a lot on the actual legal structure of the business. E.g., there are advantages to 1099-ing everyone. *And* disadvantages! (As I said, I am trying not to muddy the discussion with those issues)

I had exactly one client who *insisted* the entire "team" be collocated. Some of us lived in hotels for months at a time. None of *us* (and, by extension, the work we individually performed) benefitted from this collocation. It boiled down to the client's insecurity ("How can I be a manager if no one can see the people that I am managing?").

Since then, every relationship has been usually across state lines with no physical contact. The UPS guy and I have become

*very* friendly -- due to his frequent visits (often two or three times a week)!

Exactly. I can put together a "perfect" crew, today -- many of which have never met each other, etc. But, who would all work well with each other based on the recommendations of their peers ("I can vouch for Joe...").

The problem I see is dealing with anything beyond "steady state". I.e., what do you do when you *need* additional staff? How do you handle a "loss" (illness, death, "moving on", etc.)?

The distance makes it harder for folks to notice *personal* issues that others may be dealing with -- that could be precursors to future organizational changes (someone quitting, performance falling off, etc.). E.g., you're less likely to notice Bob is having marital problems and headed for a messy divorce if you're not socializing with him (esp after work).

And, its too easy to become "isolated TOGETHER". If you've ever worked on a really tight team, its easy to see how you can all "focus inward" on your common goal -- to the exclusion of everything around you. You *don't* interact with as many "outsiders". So, when the need for another engineer/accountant/etc arises, you're not as likely to have anyone that you are tight enough with to be able to heartily recommend ("vouch for").

But you can do that over *distributed* whiteboards. You can videoconference and simultaneously share "drawing spaces". Why force people to work at "typing speed" when you can exchange audio and video imagery in real time?

I snap several dozen digital pix each week. Friends looking at my "photo collection" never see any *people* in the imagery. One friend jokingly commented about this. I told him: "I know what Bob, Jane, Sally, Ted, etc. look like. And, if I ever have to describe any of them to someone, I just say something like 'She's a short blond I know from my work at XXXXXX' or 'He's a bike enthusiast I met while exercising at the local park'. Unless someone is looking for a 'hookup', nothing more is usually necessary.

On the other hand, if I am trying to describe the flowers on the mimosa's or how a PCB fits into an enclosure, it is impractical (and imprecise) to try to define these things textually. OTOH, snapping a photo and emailing it off makes the issue abundantly clear!"

How does a live video conference *worsen* communication skills compared to sitting in a room together? The extent of each participants' vocabulary remains unchanged. Their grammar doesn't deteriorate.

I find in many "in person" situations there are many folks who tend to sit quietly and NOT participate. Either because they aren't aggressive enough to deal with the group dynamics or aren't quick enough on their feet to fully prepare their arguments without appearing "incorrect/ill-conceived/etc.".

I can't count the number of project meetings I've been in where someone approached me afterwards to pitch an idea that they *could* have pitched 30 seconds earlier, to the group. I attribute this to them knowing that I am open to all sorts of ideas -- regardless of how off-the-wall they might sound. Or, how "unpolished" they might be. *And*, that I am willing to aggressively back (or *present* on behalf of) someone else's ideas to the group and defend them without "usurping" authorship.

One of the wins (IMO) of email is that it lets people think about their comments and organize their thoughts *before* exchanging them with others. ("Hmmm... no, that doesn't sound right. It has the following technical problems... Let me revise it.") A video conference gives folks the ability to review the "exchanges", try to refine their understanding of the issues that were *actually* presented, then fit their argument into the discussion(s) ex post factum. These can then be shared with The Group in a later "meeting" or emailed to each party for their own *individual* review or as preparation for the next meeting.

With conventional "meetings" (not "recorded"), you have none of these capabilities. Did you take ACCURATE notes? Will the other parties agree with those notes? Do they have different memories of what was "decided"/discussed? Did they understand the subtlety of a particular issue that was presented?

You're assuming old communication mechanisms. I can peer over your shoulder, remotely, *watching* you route a board and *drag* the traces that I find fault with to where I want them -- while *you* watch.

For the past ~25 years, I've never had a problem with this sort of thing. Either I laid out my own boards or had someone as technically skilled as myself do it for me (never a "CAD person").

Embedded code has *never* been a problem "with distance". It is all a function of how well *specified* the product is. Lazy bosses/employers need to be able to look over your shoulders because they were unable to decide what they actually wanted. Or, didn't trust you to make good decisions as their proxy. I frequently write code and "introduce it" to the hardware *after* the code is finished (before the hardware was stable). If you can't define how the hardware will behave, then that's not something that proximity is likely to improve!

I see problems when prototypes are scarce or rapidly evolving. I.e., you can distribute multiple copies of documents, code, etc. "for free". You can (usually) fab multiple copies of a prototype board set "inexpensively". But, if you have some precision mechanism machined from stainless steel, you are unlikely to have *two* of them!

Or, if you've hacked some new feature/design aspect onto a *particular* prototype, you care unlikely to be able to share it with others. At least not quickly or cheaply.

Many people are "imagination deficient" and *really* need to touch and feel things to internalize/understand their role or functionality. ("Hmmm... what causes this thing to pivot out of place before *that* mechanism slams into it?")

And, for some physical things, it is hard for other "eyes" to see problems without having the device in front of them to "exercise" ("Well, the 3D model *seems* to indicate that these two pieces don't interfere... that there is 0.030 clearance between them. But, is that *really* the case given all the manufacturing tolerances litering this drawing set??")

"Restricting" the audience of this (or any other portion of the design) to those closest to it (i.e., the mechanical folks to the mechanism; the electrical folks to the electronics; etc.) silently removes the ability for a cross-discipline examination to stumble on something that *should* have been "obvious" ("Um, guys, how do we changed the plugs once the engine is installed in the engine compartment??")

You've never done a dog-and-pony via a webcam? :>

The biggest benefit of "presence" is the other party can "muck with" your presentation and push it in directions that you hadn't expected. I have an innate ability to "break" (cause to misbehave) things by doing the unexpected. It is a lot harder for me to do that remotely -- and far less

*impressive* if I have to verbally *tell* you: "OK, now unplug the blue connector and type 'FOO' on the keyboard..." instead of just letting you *watch* me do it! :>

It's also not unusual in small ones! :> (especially *really* small ones)

Curiously, I tend to see things like accounting, sales, marketing, personnel, etc. -- the things that, IMO, need *less* physical co-presence -- fall under the brick and mortar roof while more "involved/interrelated" activities get farmed out!

I.e., it seems a lot easier to deal with the accounting department being off-site than the *manufacturing* group (spoken from the viewpoint of an engineer).

What I'm trying to suss out is much finer-grained than that. "Divisions" have their own structure, discipline, ethics, etc. "Inertia". I.e., the performance, reliability, efficacy, etc. of a *division* is more likely to remain constant (over time) than *individuals*. And, there are other "agents" available in that locality to resolve problems (if Bob quits, someone

*there* will replace him or hire his replacement; if Bob starts slacking off, someone *there* will see the behavior and act on it before it becomes a problem; etc.).

With "teams of consultants" (or, teams of consultant teams!), you still have the client's "big stick" in play. There is pressure from peers to "produce" lest you be thrown under the bus!

I think if you decompose the organization to the level of

*individuals* acting independently, you really elevate the quality of the personnel that you need. There's no "taskmaster" riding herd over the "group". Instead, you rely on their individual *consciences* to "not disappoint" their peers.

Bad audio, bad video, no shared whiteboard, no body language, and misses human intangibles that I think are important.

The extent of each

Some people can't brainstorm, can't play the game, because they lack the technical skills or their personality doesn't allow it. Don't invite them back.

Either because

If the atmosphere is properly managed, nobody should be inhibited by the fear of being wrong. Wrongness is an asset. And there should be no difference between a serious idea and a joke. Both jokes and wrong ideas often evolve into good stuff.

We photograph whiteboards, with titles and names and dates scribbled in the corners.

Will the

For three weeks solid?

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com   

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

Since I work for myself I do 99.9% of the work from my own office. Every now and then I meet with other team members. I've been part of several pretty complex projects and it seems to work very well.

It is very important that management keeps track on what everyone is doing and sets milestones for all team members. For some projects I also do a bit of project management. I make spreadsheets which lists the tasks to be done, remarks about the task, when it should be finished and who is reponsible. That usually works well as long as you are aware that no planning is made out of granite :-)

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

If you make sure there is a similar system on both sides it is not a problem. I do that all the time. If a customer reports a problem I ask them to describe what they do as good as they can and then I try to reproduce the problem on my test setup.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

But that only really means that the smallest "atoms" you can usefully split a company into are skill groups. In other words, the people you would care (or maybe: dare) to actually explain the problem to in the necessary detail to get to that kind of "Hold it!! You did _what_? That's exactly why it doesn't work!" moment still have to be where the coffee they had been drinking until that very moment will actually land on your shirt.

On Sat, 12 May 2012 12:56:56 -0700, Don Y wrote: [snip]

There has to be a lot of mutual respect among the players. I have worked in a some what decomposed envrionment for years. Layout person, mechanical, manufacturing expertise, all living in different locations. The mechanical guy is very good and I trust his judgement. We will have discussions about "move this here", "make that smaller", etc but if he says "this screw really has to be here", I know it really needs to be there. Layout person is damn good at what he does. I know when the gerbers come back it's going to be a work of art.

This is not to say that you might not have 3 or 4 people that have a small office or lab where they gather to work.

Where this fails is when you have people that need a lot of hand holding or dont give a rat's ass and just want a paycheck. You have to weed these folks out quickly or never let them in. It also does not work for those people that need the social interaction of a office envrionment. The people who cant wait for the next meeting invite to pop up on their calendar.

What this does is give people the ability to work at something they enjoy while at the same time living where they enjoy living. Hell, even in a small town of 600 people you can have high speed network connectivity, high enough to do skype video and more than enough for things like gotomeeting or some of the other sharing tools.

Is there a upper size limit on the complexity of the product for this concept to work? Probably, but I dont know what it is. You have car companies, aircraft companies and others that have design centers around the world all (supposedly) working together.

What I do know is it takes a different mindset and a different set of tools to make your decomposed company work. Also, in my view, it does not mean a bunch of people working "from home". Some people need an office away from their home. Perhaps there are too many distractions at home or not enough space for a lab or any other number of reasons.

You also have to deal with the current social stigma of not having a physical location, an office, a building. Every now and then I get a call from some sales guy, going to be "in the area" and wants to come by. I dont have time and could care less about going out to lunch. If you have some product that is going to rock my world, send me some info by email. I'll look at it and get back to you if I think it could be useful.

--
Chisolm

For some things, maybe.

For "mechatronic" systems with mechanical, electronic, and software content that all has to play nicely together, not so much, unless you're constantly shipping physical articles around.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Are you using a DIALUP modem?? :> I.e., you can watch "theatre quality" multimedia over a *home* internet connection. Is that not "good enough" audio/video?

That's just technology. If it hasn't been implemented already, it's trivial to do (there are no "unsolved problems" in its implementation technologies)!

Body language could be captured depending on the width of the field of view for each participant's camera. Some of the less audible cues (grumbling, shifting around in their chair, etc.) might not be. This is mainly a problem with folks who are too timid to speak their minds. Those folks would have to be coaxed to offer their *honest* opinions *regardless*.

Excluding people because they lack certain personality skills seems like A Bad Idea. I went to school with lots of folks who were "socially inept". Removing them from technical discussions would be a *huge* handicap.

(OTOH, I sure wouldn't want them out representing the company to customers, etc.! :> )

But there are people who simply *won't* participate: "He wouldn't say sh*t (even) if he had a mouthful!"

I grew up with an incredibly intelligent friend who fit this mold. Getting more than 3 words out of him was a major accomplishment! Whether this was the result of some genetic basis, the culture in which he grew up or just a reaction to the more "dominant" personalities around him, I don't know.

And, even if willing to *offer* an idea, if not willing to

*defend* and *promote* it, it can often wither before getting a fair hearing.

Equally problematic are those folks who can't *explain* their position(s). (Or, won't!) Depending on the role the person is filling, this can lead to resentment ("And just why do we have to do it *his* way??") or frustration ("OK, let's assume I *am* wrong. Explain to me *why* I am wrong; where the fault lies in my argument/understanding...")

Imagine all that -- along with the audio commentary that accompanies it as well as the *dynamics* of how the ideas evolved -- being preserved "for free". And, available at a mouseclick to anyone present (or absent!)

Surely *you* don't sit over your CAD guy's shoulders "for three weeks solid"! You check in on his progress N times daily/weekly. Or, respond to his petitions (phone calls, pages, etc.) as required. How is this any different from an alert popping up on your computer requesting advice on some aspect of his job? (If you are "away from your computer", isn't that akin to being out of the office? Does he then *telephone* you with the question? Can't he do that here, as well?)

The biggest problem, there, is making sure there is some overlap in your work schedules (*iff* there must be an interactive response). I.e., if he's an owl and you're a lark, you might have to compromise and agree to some overlap in your schedules -- even if this is only on a sporadic basis.

(Reading your words *literally*...) For "customers", you often have many devices (it's in production) so having a spare (or three) is no big deal. This isn't often the case for prototypes where only *one* might exist. Or, for products that are extremely expensive, oversized, have special environmental consequences, require special "supplies" to operate, etc.

E.g., film coating pans (used for pharmaceuticals, seed preparation, etc.) are physically large -- think of a front loading clothes dryer 6 ft wide, tall and deep. They require lots of ancillary equipment to operate (air handlers, etc.). And, lots of *power*.

Their performance varies greatly depending on the material being coated ("pills", seeds, etc.), the coating being applied, the conditions under which it is applied (air temperature, volume, moisture content, altitude, etc.) and the *process* governing the application. Furthermore, the materials involved might be hazardous, "controlled substances", proprietary, expensive, etc.

A colleague designed "dynamometers" in the past. These are pseudo-mechanical loads used to test (automobile) engines. I.e., the sort of device that your car's wheels are "pushing against" when you get your vehicle emissions tested. Not the sort of thing you are likely to be *able* to replicate test conditions on without a special facility! :>

That's a great way to put it!

But, that might *also* be too small!

You never know when someone with a different engineering discipline can expose a radically different solution to a problem than one that your "skill group" has set out to tackle. There are dualities between disciplines so equivalent solutions can exist from different approaches.

Sometimes, these solutions are incredibly humbling -- "d'oh!" moments.

At one of my first jobs, we manufactured RADAR units for maritime use. To synchronize the antenna (rotating) with the display (rotating *electronically*), a rotary encoder was placed on the antenna shaft *after* the reduction gearbox (the antenna rotates slowly to catch the echoes). This worked incredibly well and was pretty novel at the time (commercial "integrated logic" was still fairly young).

Another firm, ahem, "borrowed" the design of our RADAR. But, when we went looking under the hood, they had sited a *crude* rotary encoder on the *drive* side of the gearbox. Taking OBVIOUS advantage of the mechanical reduction that *followed* to allow them to dramatically decrease the cost and complexity of the encoder (ours was photographically produced to achieve the fineness of detail that we required -- theirs was amazingly crude!)

Agreed. You don't have to know the *material* that I know. But, you have to recognize that I *do* know it!

At the same time, I have to recognize that you are a competent and intelligent being and that I should be able to *explain* the reasons behind my decision/opinion to you to whatever degree of certainty you require. That's what I *owe* you.

[the same applies, in reverse. "Because I *said* so!" is not a valid justification! :> ]

Exactly. It should be *their* choice how best to do *their* job(s). Not something *imposed* on them from above.

Yes. I can see *creating* just such a group -- from folks with which you are already intimately familiar and comfortable. The problem I see is that it is too "comfy" an environment. There is no inherent incentive to keep probing for "new faces" to replace losses or address growth.

IMO, it's an environment that is too *ripe* with temptation for folks who are The Wrong Type. The ones who take advantage of the freedom to pursue their own interests, procrastinate, get defensive trying to rationalize their past behavior, etc.

[imagine the various sorts of compensation approaches that you might use and how each is amazingly deficient in some regard!]

Exactly. More to the point, why should *I* be forced to live somewhere just to ensure I am close to my "staff"? Why should

*they* be forced to relocate to the environment that *I* choose?

There may be industries or specific needs that require this sort of proximity. But, I suspect, there are far *fewer* than one can imagine!

Design centers imply *groups* in each location. This carries with it some peer policing, etc. As you decompose the organization to smaller elements, you rely on *self*-policing/discipline.

E.g., in the 80's, we brought an MRP system into a firm where I worked. *Everyone* in the company was trained in the system. But, not because we needed technical knowledge of the operation of the system. Rather, because we all needed to understand the underlying *philosophy* on which the system was based -- as it would now be the model by which the company conducted all of its business: *Expect* your inputs to be available when you need them; and *ensure* your outputs are available (to *your* "internal customer") when you said they would be. Build no "slop" into the system because you can now see how that multiplies up and down the chain! The incessant question was "but what happens if the stuff from ________ isn't ready when I need it?" People just couldn't wrap their heads around the inevitable reply: "it *will* be"!

Until people internalized that sort of belief system, it wouldn't work.

The same is true in the decomposed environment: people *are* depending on you and relying on you to produce what you said you would produce. That doesn't mean that schedules won't slip. "Sh*t Happens". But, this obligation to your fellows requires you to ensure this isn't a *surprise* to them!

When you are working "isolated", folks have to rely on what you explicitly *tell* them. They can't see if you are actually making progress. Or, if you are running personal errands all day, etc.

Yup. Each person's individual work requirements will dictate how they can *best* do their job.

I, for example, like being able to turn work on or off "instantly". If I am unproductive, "now", I can get up and do my grocery shopping, yardwork, take a nap, visit with neighbors, etc. There is no "investment" that I have to evaluate -- "I've already driven to the office. It would be a shame to leave and come back *later*..."

By the same token, if I wake up in the middle of the "night" with an idea, I want to be able to act on it *then* -- not at some later time/date.

I have colleagues who can't work like this. A young child might wander into their "work area" and distract them during the day. Or, phone calls from customers/vendors/etc. For them, "getting away" is the only way to be productive. And, as a consequence, when they "return home", work is *done* for the day!

Or, the folks wanting to sell you janitorial supplies, landscaping services, etc. "Um, we don't *have* any toilets. Nor 'grounds'..."